ndt-users - Re: More data points...
Subject: ndt-users list created
List archive
- From: Richard Carlson <>
- To: Phil Reese <>,
- Subject: Re: More data points...
- Date: Wed, 14 Sep 2005 13:22:41 -0400
Hi Phil;
Right off the bat I notice a big change in the reported RTT. In the first test it's 211 msec so the 64 KB buffer limits things to 2.6 Mbps. In the second test the RTT dropped to 64 msec (or 7.8 Mbps). You might want to try running some pings to various boxes to see where the delay is being introduced. It could be a time of day issue with more users causing more delays.
Rich
At 12:08 PM 9/14/2005, Phil Reese wrote:
I'm starting to think I'm calling WOLF, but here are more data points.
Last night I ran the NDT test and got the following info:
running 10s outbound test (client to server) . . . . . 667.46Kb/s
running 10s inbound test (server to client) . . . . . . 2.20Mb/s
CurCwnd: 102200
CurRwinRcvd: 65535
Duration: 10088170
Sndbuf: 223232
estimate = 52.78 based on packet size = 11Kbits, RTT = 211.03msec, and loss = 1.0E-6
The theoretical network limit is 52.78 Mbps
The NDT server has a 218.0 KByte buffer which limits the throughput to 8.07 Mbps
Your PC/Workstation has a 63.0 KByte buffer which limits the throughput to 2.36 Mbps
The network based flow control limits the throughput to 3.69 Mbps
Client Data reports link is 'T1', Client Acks report link is 'T1'
Server Data reports link is 'OC-48', Server Acks report link is 'Ethernet'
DSLReports at the same time:
Your download speed : 8652754 bps, or 8449 kbps.
A 1056.2 KB/sec transfer rate.
Your upload speed : 708497 bps, or 691 kbps.
Then this morning, early, running in the same browser window from the test of last night (first time this has given the expected results in several weeks):
running 10s outbound test (client to server) . . . . . 733.66Kb/s
running 10s inbound test (server to client) . . . . . . 7.33Mb/s
CurCwnd: 48180
CurRwinRcvd: 65535
Duration: 10065985
Sndbuf: 223232
estimate = 7.02 based on packet size = 11Kbits, RTT = 63.99msec, and loss = 6.1529E-4
The theoretical network limit is 7.02 Mbps
The NDT server has a 218.0 KByte buffer which limits the throughput to 26.61 Mbps
Your PC/Workstation has a 63.0 KByte buffer which limits the throughput to 7.81 Mbps
The network based flow control limits the throughput to 12.18 Mbps
Client Data reports link is 'Ethernet', Client Acks report link is 'T1'
Server Data reports link is 'OC-48', Server Acks report link is '10 Gig'
DSLReports at the same time:
Your download speed : 8348353 bps, or 8152 kbps.
A 1019 KB/sec transfer rate.
Your upload speed : 712115 bps, or 695 kbps.
I'm back to thinking that something odd is happening within Comcast's -- Stanford's network over time. Other thoughts?
Is there other info I might provide?
Phil
------------------------------------
Richard A. Carlson e-mail:
Network Engineer phone: (734) 352-7043
Internet2 fax: (734) 913-4255
1000 Oakbrook Dr; Suite 300
Ann Arbor, MI 48104
- More data points..., Phil Reese, 09/14/2005
- Re: More data points..., Richard Carlson, 09/14/2005
Archive powered by MHonArc 2.6.16.